Device and method for tracking sound source location by removing wind noise

• US 9,549,271 B2
• Filed: 12/27/2013
• Issued: 01/17/2017
• Est. Priority Date: 12/28/2012
• Status: Active Grant

• Current Assignee: Korea Advanced Institute of Science and Technology
• Original Assignee: Korea Advanced Institute of Science and Technology
• Inventors: Nguyen, Van Quang, Choi, Jong Suk
• Primary Examiner(s): King, Simon
• Application Number:
  US14/758,056
• Publication Number:
  US 20150358750A1
• Time in Patent Office: 1,117 Days
• US Class Current: 1/1
• CPC Class Codes: G10L 21/0208 Noise filtering
• CPC Class Codes: G10L 25/48 specially adapted for particular use
• CPC Class Codes: G10L 25/78 Detection of presence or absence of voice signals switching of direction of transmission by voice frequency in two-way loud-speaking telephone systems H04M9/10
• CPC Class Codes: H04N 7/188 Capturing isolated or intermittent images triggered by the occurrence of a predetermined event, e.g. an object reaching a predetermined position signal generation from motion picture films H04N5/253
• CPC Class Codes: H04R 2410/07 Mechanical or electrical reduction of wind noise generated by wind passing a microphone
• CPC Class Codes: H04R 29/00 Monitoring arrangements; Testing arrangements for hearing aids H04R25/30; detection of loudspeaker connection H04R5/04; sound-field adaptation dependent on speaker detection H04S7/308
• CPC Class Codes: H04R 3/002 Damping circuit arrangements for transducers, e.g. motional feedback circuits
• CPC Class Codes: H04R 3/005 for combining the signals of two or more microphones specially adapted for hearing aids H04R25/407

Abstract

Disclosed are a device and a method for tracking a sound source location by removing wind noise. The disclosed method for tracking a sound source by removing wind noise comprises the steps of: performing a Fourier transform for input signals of a time domain received by a plurality of microphones; detecting a first sound source section on the basis of an average power value of the Fourier-transformed input signals; detecting a second sound source section from which wind noise has been removed, on the basis of the first sound source section and a difference of power values of the input signals received by the microphones; and detecting a location of a sound source on the basis of the second sound source section and a phase difference of the input signals received by the plurality of microphones.

Claims

1. A method for localizing a sound source by removing wind noise, comprising:

performing a Fourier transform for input signals of a time domain received by a plurality of microphones;

detecting a first sound source section on the basis of an average power value of the Fourier-transformed input signals;

detecting a second sound source section from which wind noise has been removed, on the basis of the first sound source section and a difference of power values of the input signals received by the microphones; and

detecting a location of a sound source on the basis of the second sound source section and a phase difference of the input signals received by the plurality of microphones.

2. The method for localizing a sound source by removing wind noise according to claim 1, further comprising:

calculating reliability for the detection of a location of the sound source,

wherein the detecting of a first sound source section further includes detecting the first sound source section only when the first sound source section is continuously formed.

3. The method for localizing a sound source by removing wind noise according to claim 1,

wherein in the detecting of a first sound source section, the average power value of the input signals is determined using Equation 1 below, and

wherein the first sound source section is determined using Equation 2 below:

4. The method for localizing a sound source by removing wind noise according to claim 3,

wherein the f_min is 300 Hz, and the f_max is 3.4 kHz.

5. The method for localizing a sound source by removing wind noise according to claim 3,

wherein in the detecting of a second sound source section, a cross-channel power difference is calculated using Equation 3 below, and

wherein the presence of wind noise is determined using Equation 4 below:

6. A device for localizing a sound source by removing wind noise, comprising:

a plurality of microphones;

a Fourier transform unit configured to performing a Fourier transform for input signals of a time domain received by the plurality of microphones;

a first sound source section detecting unit configured to detect a first sound source section on the basis of an average power value of the Fourier-transformed input signals;

a second sound source section detecting unit configured to detect a second sound source section from which wind noise has been removed, on the basis of the first sound source section and a difference of power values of the input signals received by the microphones; and

a sound source position detecting unit configured to detect a location of a sound source on the basis of the second sound source section and a phase difference of the input signals received by the plurality of microphones.

7. The device for localizing a sound source by removing wind noise according to claim 6, further comprising:

a reliability calculating unit configured to calculate reliability for the detection of a location of the sound source,

wherein the first sound source section detecting unit detects the first sound source section only when the first sound source section is continuously formed.

8. The device for localizing a sound source by removing wind noise according to claim 6,

wherein the first sound source section detecting unit determines the average power value of the input signals by using Equation 1 below, and

wherein the first sound source section detecting unit determines the first sound source section by using Equation 2 below:

9. The device for localizing a sound source by removing wind noise according to claim 8,

wherein the f_min is 300 Hz, and the f_max is 3.4 kHz.

10. The device for localizing a sound source by removing wind noise according to claim 8,

wherein the second sound source section detecting unit calculates a cross-channel power difference by using Equation 3 below, and

wherein the second sound source section detecting unit determines the presence of wind noise by using Equation 4 below: